Related Experiment Videos
Artificial neural network: border detection in echocardiography
Eduardo Jyh Herng Wu1, Márcio Luiz De Andrade, Denys E Nicolosi
1University of Campinas, Campinas, SP, Brazil. wuherng@yahoo.com.br
Insights
This study introduces an artificial neural network (ANN) method for precise left ventricle border detection in noisy echocardiography images. The ANN approach proved more effective than manual delineation by specialists.
Area of Science:
- Medical Imaging
- Cardiology
- Artificial Intelligence
Background:
- Echocardiography is a vital, non-invasive tool for assessing left ventricle function.
- Image noise in echocardiograms significantly hinders accurate delineation of cardiac structures.
- Existing segmentation methods struggle with the inherent noise in echocardiographic data.
Purpose of the Study:
- To develop and evaluate an artificial neural network (ANN) based method for accurate left ventricle border segmentation in echocardiography.
- To overcome the limitations of image noise in echocardiographic analysis.
- To improve the efficiency and accuracy of left ventricle border detection.
Main Methods:
- Utilized artificial neural networks (ANNs), known for noise resilience, for image segmentation.
- Implemented operator-selected regions of interest to optimize processing time.
- Employed neighborhood and gradient search techniques to refine contour tracing.
Main Results:
- The ANN method demonstrated high efficiency in detecting left ventricle borders in echocardiography images.
- Results showed superior performance compared to manual delineation by medical specialists.
- Proper selection of analysis areas and central points is crucial for optimal results.
Conclusions:
- Artificial neural networks offer a robust solution for segmenting noisy echocardiography images.
- The proposed method enhances the accuracy and reliability of left ventricle border identification.
- This technique holds potential for improving diagnostic capabilities in echocardiography.
Abstract:
Being non-invasive and low cost, the echocardiography has become a diagnostic technique largely applied for the determination of the left ventricle systolic and diastolic volumes, which are used indirectly to calculate the left ventricle ejection volume, the cardiac cavities muscular contraction, the regional ejection fraction, the myocardial thickness, and the ventricular mass, etc. However, the image is very noisy, which renders the delineation of the borders of the left ventricle very difficult. While there are many techniques image segmentation, this work chooses the artificial neural network (ANN) since it is not very sensitive to noise. In order to reduce the processing time, the operator selects the region of interest where the neural network will identify the borders. Neighborhood and gradient search techniques are then employed to link the points and the left ventricle contour is traced. The present method has been efficient in detecting the left ventricle borders echocardiography images compared to those whose borders were delineated by the specialists. For good results, it is important to choose properly the areas to be analyzed and the central points of these areas.
